Pith. sign in

REVIEW 4 major objections 6 minor 38 references

High-resolution rapid-scanning Fourier-transform spectroscopy of ultracold atoms

T0 review · 4 major / 6 minor · reviewed 2026-08-08 · deepseek-v4-flash

Pith's one-line read Rapid-scanning Fourier-transform spectroscopy of ultracold atoms reaches 250 MHz spectral resolution, a tenfold improvement over step-wise scanning.

desk verdict Believable 10x resolution gain for rapid-scanning PM FT spectroscopy on ultracold atoms, but the headline 250 MHz linewidth rests on one unquantified measurement and an unverified delay-axis calibration. read the letter →

arxiv 2502.08446 v2 pith:VARB6A6H submitted 2025-02-12 physics.atom-ph

classification physics.atom-ph
keywords femtosecondspectroscopyFourier-transformultracoldatomsacousto-opticphasemodulationrapidscanningactiondetectionmagneto-opticaltraplithium
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

Femtosecond pulses naturally combine broad bandwidth with poor spectral resolution, which has kept ultrafast nonlinear spectroscopy away from the narrow lines of ultracold atoms. This paper shows that a continuous "rapid-scanning" version of phase-modulated Fourier-transform interferometry closes that gap. Using a stabilized diode laser to reconstruct the interferometer delay in real time, the authors measure a fluorescence-detected spectrum of laser-cooled lithium atoms with 250 MHz full width—more than ten times sharper than the 2.9 GHz they had reached with step-wise scanning. The same scheme resolves 1.17 GHz-spaced hyperfine doublets in cesium vapor that step-wise scanning buries in noise, while covering the delay range about 350 times faster. If the result holds, femtosecond coherent and multidimensional spectroscopy can at last exploit the narrow linewidths of ultracold quantum systems.

What carries the argument

The load-bearing mechanism is the real-time delay reconstruction of a rapidly scanned, phase-modulated Mach-Zehnder interferometer. A folded 300 mm mechanical delay line provides a 3.9 ns scan range; two phase-locked acousto-optic modulators generate a 5 kHz beat note $\Omega_{12}$; and a stabilized continuous-wave diode laser provides a reference interferogram $S_{\mathrm{ref}}(t)$ whose Hilbert-demodulated phase is converted into the delay axis $\tau(t) = (c_{\mathrm{ref}}/c_{\mathrm{sam}})\,\varphi_{\mathrm{ref}}(t)/\omega_{\mathrm{ref}}$. This turns an unknown mechanical sweep into a precisely known optical phase, enables 1 fs binning of the interferogram, and moves the phase-noise scale from the sample frequency $\omega_{\mathrm{sam}}$ to the difference frequency $\Delta\omega=\omega_{\mathrm{sam}}-\omega_{\mathrm{ref}}$, which is what makes sub-gigahertz line shapes stable.

What would settle it

Record the same lithium transition with the delay axis calibrated independently—for example by a second reference laser at a different wavelength or by an optical frequency comb—and compare the reconstructed line position and width to the 250 MHz result; if the real-time $\tau(t)$ reconstruction is exact, the two calibrations must agree to well below the instrument response, whereas any error in the assumption $c_{\mathrm{ref}}/c_{\mathrm{sam}}=1$ or in the diode-laser phase would show up as a line shift or broadening.

Watch

Extended reading notes

Core claim

The central discovery is that the resolution limit of phase-modulated femtosecond Fourier-transform spectroscopy is set less by the length of the delay scan than by how accurately the delay axis is known, and that continuous rapid scanning with real-time phase tracking removes that limit. In the rapid-scanning scheme the delay stage is swept continuously while a temperature- and current-stabilized diode laser at frequency $\omega_{\mathrm{ref}}$ is sent through the same interferometer; its phase $\varphi_{\mathrm{ref}}(t)$, recovered against the AOM beat $M_{12}(t)$, gives the instantaneous delay $\tau(t) = (c_{\mathrm{ref}}/c_{\mathrm{sam}})\,\varphi_{\mathrm{ref}}(t)/\omega_{\mathrm{ref}}$, with effective sub-attosecond steps that are then binned to 1 fs. This active correction suppresses stage irregularities and reduces the phase-noise sensitivity from $\omega_{\mathrm{sam}}\,\delta\tau$ to the much smaller $\Delta\omega\,\delta\tau$, where $\Delta\omega=\omega_{\mathrm{sam}}-\omega_{\mathrm{ref}}$. Applied to laser-cooled Li atoms, the method resolves the $2\,^2P_{3/2}\to 3\,^2S_{1/2}$ transition with a 250 MHz FWHM, matching the 256 MHz instrument function of the 3.9 ns scan range—a tenfold improvement over the earlier step-wise result.

Load-bearing premise

The 250 MHz result stands on the assumption that the delay between the two femtosecond pulses can be read off exactly from the phase of the separate reference laser at every moment of the scan; if that laser's frequency drifts or the two colours do not travel at the same speed through the interferometer, the corrected time axis bends and the measured line broadens.

Editorial extensions

If this is right

  • Ultracold-atom femtosecond spectra can now resolve sub-gigahertz features, so the narrow Doppler width of cold samples no longer goes unused.
  • For the same delay range, the rapid scan takes about 350 times less measurement time than step-wise scanning, reducing drift and enabling faster parameter surveys.
  • The $1/\sqrt{N}$ statistical-noise scaling is preserved, so scan speed can be chosen to trade acquisition time against signal-to-noise ratio without degrading resolution.
  • Because the phase-modulation detection is unchanged, the rapid-scanning scheme transfers directly to multi-pulse and multidimensional nonlinear spectroscopy of ultracold quantum systems.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • If the diode-laser phase is indeed the only remaining noise source, then borrowing frequency-comb techniques should push the resolution below 100 MHz without changing the interferometer, since the delay axis would then be known to comb-line accuracy.
  • The same real-time delay reconstruction could be applied to each interferometer in a multi-pulse 2D or multidimensional scheme, so sub-gigahertz-resolved coherent spectra of cold molecules or Rydberg systems become a plausible next target.
  • A direct check the paper does not report is to scan the same Li transition at different stage speeds; if the reconstruction is exact, the reconstructed delay axes and line widths should coincide, whereas any residual velocity-dependent effect would appear as a scan-speed-dependent broadening.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 6 minor

Summary. The manuscript reports a rapid-scanning implementation of phase-modulated Fourier-transform spectroscopy and compares it with the step-wise scanning version. In a Cs vapor cell, the rapid-scanning mode resolves the D1 hyperfine doublets with a 1.3 ns scan range (quoted instrument response 0.8 GHz) in about 130 s, whereas the step-wise scan took about 13 h and did not resolve the doublets. For laser-cooled Li atoms, the authors record a 3.9 ns scan and report an action-detected line with a FWHM of 250 MHz, which they state is more than a tenfold improvement over the 2.9 GHz resolution of their previous MOT experiment. The delay axis is reconstructed in real time from the phase of a temperature- and current-stabilized, but not frequency-locked, 384 THz diode laser, assuming cref/csam = 1.

Significance. If the calibration and linewidth figures hold, this is a valuable technical step: it extends phase-modulated Fourier-transform spectroscopy into the sub-gigahertz regime for ultracold samples, with acquisition times that are orders of magnitude shorter than step-wise scanning. The explicit comparison of step-wise and rapid-scanning modes, the SNR-versus-N scaling check in Fig. 2e, and the use of the Cs hyperfine structure as a coarse frequency benchmark are useful internal validations. The method is described with enough detail to be reproduced, although no machine-checked proofs or code are provided.

major comments (4)
  1. [Experimental setup] The delay axis is reconstructed as τ(t)=cref/csam·φref(t)/ωref with cref/csam=1, using a diode laser that is temperature- and current-stabilized but not frequency-locked. Since the reference (384.001 THz) and sample (368.81 THz for Li, 335.116 THz for Cs) wavelengths differ, air dispersion introduces a constant scale error of order 100 MHz in the apparent Li line-center frequency, and any drift of the diode laser during the approximately 130-200 s scan produces a time-dependent phase error that is not corrected by the active tracking. For example, a linear diode-laser drift of about 40 MHz over the scan leads to a phase error near 1 rad at the end of the 3.9 ns scan, which is sufficient to broaden or distort the claimed 250 MHz line. The authors provide no measurement of the diode-laser frequency stability and no independent calibration of the delay axis at the 100 MHz level; the Cs hyperfine splitting of 1.17 GHz is too coarse for this purpose. This issue is load-bearing for the central resolution claim and should be addressed with a stability measurement and/or a calibrated frequency reference.
  2. [Results (Li measurement)] The text equates the instrument response function with 1/T = 256 MHz for T = 3.9 ns and then reports a measured FWHM of 250 MHz in 'good agreement'. For a rectangular time window, the Fourier-transform line shape is a sinc function whose FWHM is approximately 1.206/T ≈ 309 MHz, not 1/T ≈ 256 MHz. The reported 250 MHz FWHM is thus smaller than the minimum FWHM for the stated scan range under the standard boxcar-window model, which is internally inconsistent. The authors should state the exact line-shape model including any apodization or windowing, report the measured FWHM with its uncertainty, and compare it with the correct theoretical line-shape width.
  3. [Results (Li measurement)] The 250 MHz value is presented as a single number without an uncertainty, number of repetitions, or a noise-floor estimate for the Li measurement. Because the claim rests on a single spectrum, the reader cannot distinguish the nominal transform limit from a line that is broadened by delay-axis errors or narrowed by a processing artifact. Repeated scans with a statistical uncertainty on the FWHM, or at least an estimate based on the noise level and a line-shape fit, are needed to support the headline resolution claim.
  4. [Results (Cs comparison)] The comparison in Fig. 2 is not fully controlled: the step-wise and rapid-scanning data differ in acquisition time (13 h vs 130 s), step size (30 fs with aliasing vs continuous sampling with 1 fs binning), and SNR (500 vs 55). The conclusion that rapid-scanning is superior for high-resolution measurements is plausible, but the step-wise result may be degraded by the long acquisition time and the aliasing/unwrapping procedure rather than by the step-wise principle itself. A controlled comparison, for example with the same scan range and step size or with active delay correction disabled, would strengthen the central comparison claim.
minor comments (6)
  1. [Throughout] There are several typos: 'reslution' should be 'resolution', 'more then' should be 'more than', 'setp-wise' should be 'step-wise', and 'measurment' should be 'measurement'.
  2. [Experimental setup] The transition notation '22P3/2 32S1/2' should be written in standard spectroscopic form, e.g., 2^2P_{3/2} → 3^2S_{1/2}, to avoid ambiguity.
  3. [Results (Fig. 2e)] The caption and text describe the SNR as a function of scanning speed and of the number of data points, but the relationship between these two axes via the scanning speed and sampling rate should be stated explicitly.
  4. [Data availability] The data availability statement says the data 'will be made available' on Zenodo; the final version should include a link or DOI.
  5. [Results (Li measurement)] The manuscript mentions zero-padding but does not state whether any apodization window is applied; this information is important for interpreting the reported FWHM and its comparison with the instrument response.
  6. [Results] The sentence comparing the data quality with 'an identical previous experiment' (Ref. [9]) should specify what was identical, since the previous experiment used a different scan range and possibly different conditions.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the 250 MHz resolution is benchmarked against the theoretical instrument response and atomic line spacings, not reduced to a fit or self-citation.

full rationale

The paper's central resolution claim is not circular. The 250 MHz FWHM is an independently measured width of the Li 2^2P_3/2 -> 3^2S_1/2 line, compared directly to the theoretical instrument response of 256 MHz obtained from the 3.9 ns scan range; it is also consistent with the Cs hyperfine doublet spacing of 1.17 GHz used as a frequency-axis check in Fig. 2. The delay-axis reconstruction tau(t) = c_ref/c_sam * phi_ref(t)/omega_ref is a calibration procedure, not a fit; the reference phase phi_ref is recorded in real time and the sample interferogram is demodulated with the same reference, but the measured linewidth is not forced by construction. The comparison to the authors' prior MOT experiment (Ref. 9) is an experimental baseline rather than a fitted parameter or imported uniqueness theorem; it is externally falsifiable and does not carry the derivation. The rapid-scanning method itself is attributed to Refs. [14,15], which are independent of the present authors. Concerns about diode-laser drift and the assumption c_ref/c_sam = 1 are calibration and correctness risks, not circularity, because no quantity in the derivation is defined in terms of the claimed result. Thus no circular step is identified.

Assumptions & free parameters 0 free parameters · 4 assumptions · 0 invented entities

The central claim rests on three types of background assumptions rather than on fitted parameters: the reference diode laser is stable and tracks the true delay, pathlength noise is common-mode between sample and reference arms, and the measured Li line is dominated by the instrument response. No free parameters are fitted to the reported spectra; the 1 fs binning and 83.3 kHz sampling are acquisition choices, not fitted values. No invented entities are introduced.

assumptions (4)
  • domain assumption The ratio of the speed of light at the diode reference wavelength and at the fs laser wavelength is cref/csam = 1.
    This is stated explicitly in the delay reconstruction formula τ(t)=cref/csam·φref(t)/ωref; if chromatic dispersion in air or optics causes the two speeds to differ, the reconstructed delay axis is systematically scaled and the reported frequency scale shifts.
  • domain assumption Pathlength fluctuations δτ are common-mode between the sample and reference interferometer arms.
    Eq. (2) derives the demodulated interferogram as cos[Δω(τ+δτ)], which is the basis of the phase-noise advantage; it holds only if sample and reference signals experience the same delay jitter.
  • domain assumption The diode reference laser remains frequency-stable to much less than the 0.8 GHz instrument resolution over the full scan.
    The text states a clean Sref over the whole scan range requires frequency stability much better than 0.8 GHz, but no measurement of the diode laser stability is provided; drift would directly broaden the reconstructed delay axis.
  • domain assumption The 250 MHz Li linewidth is dominated by the Fourier-transform instrument response, with negligible contribution from the atomic transition width.
    The FWHM is compared with the theoretical 256 MHz transform limit without deconvolution; if the transition itself has a larger width or if phase errors mix dispersive and absorptive components, the linewidth would no longer be a clean resolution benchmark.

how reviews work

0 comments
Cite this review

Pith. "Pith review of High-resolution rapid-scanning Fourier-transform spectroscopy of ultracold atoms." pith.science (2026). https://pith.science/paper/VARB6A6H

@misc{pith2026250208446,
  author       = {Pith},
  title        = {Pith review of: High-resolution rapid-scanning Fourier-transform spectroscopy of ultracold atoms},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/VARB6A6H}},
  note         = {Machine review of arXiv:2502.08446}
}
read the original abstract

Femtosecond interferometry combined with acousto-optical phase modulation is an effective approach to implement various types of coherent nonlinear and multidimensional spectroscopy schemes. The high sensitivity of this method has recently enabled the study of highly dilute gaseous and ultracold quantum systems for which the attainable spectral resolution is of particular interest. Here, we directly compare the performance and spectral resolution between two experimental implementations, that are step-wise and continuous rapid scanning of the underlying Fourier transform interferometers. We show the performance advantage of the rapid-scanning approach and demonstrate a spectral resolution of 250 MHz in the spectroscopy of laser-cooled Li atoms. This is a 10-fold resolution improvement compared to previous experiments.

Figures

Figures reproduced from arXiv: 2502.08446 by the authors.

Figure 1
Figure 1. Experimental scheme: (a) optical setup, (b) signal pro￾cessing. See main text for details. compared to the previous experiments in a MOT [9]. Experimental setup The PM technique has been described in detail before [12], including the rapid-scanning method [14]. We therefore restrict our discussion to the most essential facts. In most FT interferometers the optical absorption is measured. In contrast, the PM approach… view at source ↗
Figure 2
Figure 2. FT spectra of Cs vapor recorded with the step-wise (a) and rapid-scanning (b) of the optical interferometer (νCs = 335.116 THz). Labels indicate the hyperfine transi￾tions. (c,d) shows a zoom-out of the data of (a,b). The area dominated by white noise is marked in grey. (a-d) show the absolute value of the Fourier spectrum. (e) SNR of the rapid￾scanning measurements as a function of scanning speed (top axis) and the… view at source ↗
Figure 3
Figure 3. FT spectrum (real part) of laser-cooled Li atoms. Labels indicate the atomic resonance and the FWHM of the spectral line. noise (Fig. 2c,d). Apparently, the step-wise scanning scheme pro￾duces a smaller statistical noise floor. For a quantitative analysis, we computed the overall SNR of the Fourier spectra by dividing the maximum amplitude of the spectrum by the RMS value of the noise floor evaluated in the gray are… view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

38 extracted references · 38 canonical work pages

  1. [1]

    Georgescu, S

    I. Georgescu, S. Ashhab, and F . Nori, Rev. Mod. Phys.86, 153 (2014). Publisher: American Physical Society

  2. [2]

    V. O. Lorenz and S. T. Cundiff, Phys. Rev. Lett. 95, 163601 (2005)

  3. [3]

    Bangert, F

    U. Bangert, F . Stienkemeier, and L. Bruder, Nat Commun 13, 3350 (2022). Number: 1 Publisher: Nature Publishing Group

  4. [4]

    X. Dai, M. Richter, H. Li, et al., Phys. Rev. Lett. 108, 193201 (2012)

  5. [5]

    D. M. Jonas, Annu. Rev. Phys. Chem. 54, 425 (2003)

  6. [6]

    Bruder, U

    L. Bruder, U. Bangert, M. Binz, et al., Nat Commun 9, 4823 (2018)

  7. [7]

    Roeding and T

    S. Roeding and T. Brixner, Nat Commun 9, 2519 (2018)

  8. [8]

    Bruder, A

    L. Bruder, A. Eisfeld, U. Bangert, et al., Phys. Chem. Chem. Phys. 21, 2276 (2019)

Show all 38 references
  1. [9]

    Landmesser, T

    F . Landmesser, T. Sixt, K. Dulitz,et al., Opt. Lett., OL 48, 473 (2023). Publisher: Optica Publishing Group

  2. [10]

    Liang, L

    D. Liang, L. S. Rodriguez, H. Zhou, et al., Opt. Lett. 47, 6452 (2022). Publisher: Optica Publishing Group

  3. [11]

    F . D. Fuller and J. P . Ogilvie, Annu. Rev. Phys. Chem.66, 667 (2015)

  4. [12]

    P . F . Tekavec, G. A. Lott, and A. H. Marcus, J. Chem. Phys.127, 214307 (2007)

  5. [13]

    D. Uhl, U. Bangert, L. Bruder, and F . Stienkemeier, Optica 8, 1316 (2021). Publisher: Optical Society of America

  6. [14]

    Agathangelou, A

    D. Agathangelou, A. Javed, F . Sessa, et al. , J. Chem. Phys. 155, 094201 (2021). Publisher: American Institute of Physics

  7. [15]

    T. M. Autry, G. Moody, J. Fraser, et al., Optica 6, 735 (2019)

  8. [16]

    K. J. Karki, J. R. Widom, J. Seibt, et al., Nat Commun 5, 5869 (2014)

  9. [17]

    Wituschek, L

    A. Wituschek, L. Bruder, E. Allaria, et al., Nat Commun 11, 883 (2020)

  10. [18]

    Bruder, U

    L. Bruder, U. Bangert, M. Binz, et al., J. Phys. B: At. Mol. Opt. Phys. 52, 183501 (2019)

  11. [19]

    Lomsadze, B

    B. Lomsadze, B. C. Smith, and S. T. Cundiff, Nat Photon 12, 676 (2018). Letter 5 FULL REFERENCES

  12. [20]

    Quantum simulation,

    I. Georgescu, S. Ashhab, and F . Nori, “Quantum simulation,” Rev. Mod. Phys. 86, 153–185 (2014). Publisher: American Physical Society

  13. [21]

    Non-Markovian Dynamics in a Dense Potassium Vapor,

    V. O. Lorenz and S. T. Cundiff, “Non-Markovian Dynamics in a Dense Potassium Vapor,” Phys. Rev. Lett.95, 163601 (2005)

  14. [22]

    High-resolution two- dimensional electronic spectroscopy reveals the homogeneous line profile of chromophores solvated in nanoclusters,

    U. Bangert, F . Stienkemeier, and L. Bruder, “High-resolution two- dimensional electronic spectroscopy reveals the homogeneous line profile of chromophores solvated in nanoclusters,” Nat Commun 13, 3350 (2022). Number: 1 Publisher: Nature Publishing Group

  15. [23]

    Two-Dimensional Double-Quantum Spectra Reveal Collective Resonances in an Atomic Vapor,

    X. Dai, M. Richter, H. Li, et al., “Two-Dimensional Double-Quantum Spectra Reveal Collective Resonances in an Atomic Vapor,” Phys. Rev. Lett. 108, 193201 (2012)

  16. [24]

    Two-Dimensional Femtosecond Spectroscopy,

    D. M. Jonas, “Two-Dimensional Femtosecond Spectroscopy,” Annu. Rev. Phys. Chem. 54, 425–463 (2003)

  17. [25]

    Coherent multidimensional spectroscopy of dilute gas-phase nanosystems,

    L. Bruder, U. Bangert, M. Binz, et al. , “Coherent multidimensional spectroscopy of dilute gas-phase nanosystems,” Nat Commun 9, 4823 (2018)

  18. [26]

    Coherent two-dimensional electronic mass spectrometry,

    S. Roeding and T. Brixner, “Coherent two-dimensional electronic mass spectrometry,” Nat Commun9, 2519 (2018)

  19. [27]

    Delocalized excitons and interaction effects in extremely dilute thermal ensembles,

    L. Bruder, A. Eisfeld, U. Bangert, et al. , “Delocalized excitons and interaction effects in extremely dilute thermal ensembles,” Phys. Chem. Chem. Phys. 21, 2276–2282 (2019)

  20. [28]

    Two-dimensional electronic spectroscopy of an ultracold gas,

    F . Landmesser, T. Sixt, K. Dulitz,et al., “Two-dimensional electronic spectroscopy of an ultracold gas,” Opt. Lett., OL48, 473–476 (2023). Publisher: Optica Publishing Group

  21. [29]

    Optical two-dimensional coherent spectroscopy of cold atoms,

    D. Liang, L. S. Rodriguez, H. Zhou, et al., “Optical two-dimensional coherent spectroscopy of cold atoms,” Opt. Lett.47, 6452–6455 (2022). Publisher: Optica Publishing Group

  22. [30]

    Experimental Implementations of Two- Dimensional Fourier Transform Electronic Spectroscopy,

    F . D. Fuller and J. P . Ogilvie, “Experimental Implementations of Two- Dimensional Fourier Transform Electronic Spectroscopy,” Annu. Rev. Phys. Chem. 66, 667–690 (2015)

  23. [31]

    Fluorescence-detected two-dimensional electronic coherence spectroscopy by acousto-optic phase modulation,

    P . F . Tekavec, G. A. Lott, and A. H. Marcus, “Fluorescence-detected two-dimensional electronic coherence spectroscopy by acousto-optic phase modulation,” J. Chem. Phys. 127, 214307–214307–21 (2007)

  24. [32]

    Coherent optical 2D photoelectron spectroscopy,

    D. Uhl, U. Bangert, L. Bruder, and F . Stienkemeier, “Coherent optical 2D photoelectron spectroscopy,” Optica8, 1316–1324 (2021). Publisher: Optical Society of America

  25. [33]

    Phase-modulated rapid-scanning fluorescence-detected two-dimensional electronic spec- troscopy,

    D. Agathangelou, A. Javed, F . Sessa, et al. , “Phase-modulated rapid-scanning fluorescence-detected two-dimensional electronic spec- troscopy,” J. Chem. Phys.155, 094201 (2021). Publisher: American Institute of Physics

  26. [34]

    Single-scan acquisition of multiple multidimensional spectra,

    T. M. Autry, G. Moody, J. Fraser, et al. , “Single-scan acquisition of multiple multidimensional spectra,” Optica 6, 735 (2019)

  27. [35]

    Coherent two-dimensional pho- tocurrent spectroscopy in a PbS quantum dot photocell,

    K. J. Karki, J. R. Widom, J. Seibt,et al., “Coherent two-dimensional pho- tocurrent spectroscopy in a PbS quantum dot photocell,” Nat Commun 5, 5869 (2014)

  28. [36]

    Tracking attosecond elec- tronic coherences using phase-manipulated extreme ultraviolet pulses,

    A. Wituschek, L. Bruder, E. Allaria, et al., “Tracking attosecond elec- tronic coherences using phase-manipulated extreme ultraviolet pulses,” Nat Commun 11, 883 (2020)

  29. [37]

    Coherent multidimensional spectroscopy in the gas phase,

    L. Bruder, U. Bangert, M. Binz, et al. , “Coherent multidimensional spectroscopy in the gas phase,” J. Phys. B: At. Mol. Opt. Phys. 52, 183501 (2019)

  30. [38]

    Tri-comb spectroscopy,

    B. Lomsadze, B. C. Smith, and S. T. Cundiff, “Tri-comb spectroscopy,” Nat Photon 12, 676 (2018)

Pith tools

Reviewed August 8, 2026 · model on record in the stance chip above.